Unpaired Image-to-Image Translation via a Self-Supervised Semantic Bridge

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Jiaming, Petersen, Felix, Gao, Yunhe, Zhang, Yabin, Kim, Hyojin, Chaudhari, Akshay S., Sun, Yu, Ermon, Stefano, Gatidis, Sergios
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914337688387584
author Liu, Jiaming
Petersen, Felix
Gao, Yunhe
Zhang, Yabin
Kim, Hyojin
Chaudhari, Akshay S.
Sun, Yu
Ermon, Stefano
Gatidis, Sergios
author_facet Liu, Jiaming
Petersen, Felix
Gao, Yunhe
Zhang, Yabin
Kim, Hyojin
Chaudhari, Akshay S.
Sun, Yu
Ermon, Stefano
Gatidis, Sergios
contents Adversarial diffusion and diffusion-inversion methods have advanced unpaired image-to-image translation, but each faces key limitations. Adversarial approaches require target-domain adversarial loss during training, which can limit generalization to unseen data, while diffusion-inversion methods often produce low-fidelity translations due to imperfect inversion into noise-latent representations. In this work, we propose the Self-Supervised Semantic Bridge (SSB), a versatile framework that integrates external semantic priors into diffusion bridge models to enable spatially faithful translation without cross-domain supervision. Our key idea is to leverage self-supervised visual encoders to learn representations that are invariant to appearance changes but capture geometric structure, forming a shared latent space that conditions the diffusion bridges. Extensive experiments show that SSB outperforms strong prior methods for challenging medical image synthesis in both in-domain and out-of-domain settings, and extends easily to high-quality text-guided editing.
format Preprint
id arxiv_https___arxiv_org_abs_2602_16664
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Unpaired Image-to-Image Translation via a Self-Supervised Semantic Bridge
Liu, Jiaming
Petersen, Felix
Gao, Yunhe
Zhang, Yabin
Kim, Hyojin
Chaudhari, Akshay S.
Sun, Yu
Ermon, Stefano
Gatidis, Sergios
Computer Vision and Pattern Recognition
Adversarial diffusion and diffusion-inversion methods have advanced unpaired image-to-image translation, but each faces key limitations. Adversarial approaches require target-domain adversarial loss during training, which can limit generalization to unseen data, while diffusion-inversion methods often produce low-fidelity translations due to imperfect inversion into noise-latent representations. In this work, we propose the Self-Supervised Semantic Bridge (SSB), a versatile framework that integrates external semantic priors into diffusion bridge models to enable spatially faithful translation without cross-domain supervision. Our key idea is to leverage self-supervised visual encoders to learn representations that are invariant to appearance changes but capture geometric structure, forming a shared latent space that conditions the diffusion bridges. Extensive experiments show that SSB outperforms strong prior methods for challenging medical image synthesis in both in-domain and out-of-domain settings, and extends easily to high-quality text-guided editing.
title Unpaired Image-to-Image Translation via a Self-Supervised Semantic Bridge
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.16664